Optical learning neurochip with internal analog memory.

An optical neurochip with learning capability and a memory function is reported for the first time, to our knowledge. The neurochip is a three-dimensional optoelectronic integrated circuit consisting of a light-emitting diode array and a variable-sensitivity photodetector array. The principle of operation and the fundamental characteristics of the neurochip are described in detail. The synaptic weights, which, are acquired through learning, are stored in the variable-sensitivity photodetector as the detection sensitivity with analog values. Both the positive and the negative synaptic weights are memorized with one variable-sensitivity photodetector element by changing the polarity of the electric signal applied to the photodetector. A storage time of ~20 min. was obtained. With a fabricated optical neurochip that had 32 neurons and 32 x 32 synapses, experiments of on-chip learning for pattern classification were performed successfully.

[1]  Resonant cavity enhanced AlGaAs/GaAs heterojunction phototransistors with an intermediate InGaAs layer in the collector , 1990 .

[2]  K Kojima,et al.  Optical neurochip based on a three-layered feed-forward model. , 1990, Optics letters.

[3]  Kazuo Kyuma,et al.  GaAs/AlGaAs Optical Interconnection Chip for Neural Network , 1989 .

[4]  D. Day,et al.  Impact ionization of deep traps in semi‐insulating GaAs substrates , 1990 .

[5]  D Psaltis,et al.  Optical information processing based on an associative-memory model of neural nets with thresholding and feedback. , 1985, Optics letters.

[6]  M. Ito,et al.  Low dark current GaAs metal-semiconductor-metal (MSM) photodiodes using WSi x contacts , 1986 .

[7]  M. Takahashi,et al.  Optical neurochip with learning capability , 1992, IEEE Photonics Technology Letters.

[8]  J. Ohta,et al.  GaAs/AlGaAs optical synaptic interconnection device for neural networks. , 1989, Optics letters.

[9]  J Ohta,et al.  Dynamic optical neurochip using variable-sensitivity photodiodes. , 1991, Optics letters.

[10]  N H Farhat,et al.  Optoelectronic analogs of self-programming neural nets: architecture and methodologies for implementing fast stochastic learning by simulated annealing. , 1987, Applied optics.

[11]  J. Ohta,et al.  Variable sensitivity photodetector for optical neural networks , 1991 .

[12]  M. Lampert,et al.  Current injection in solids , 1970 .

[13]  D Psaltis,et al.  Optical implementation of the Hopfield model. , 1985, Applied optics.

[14]  S. M. Sze,et al.  Current transport in metal-semiconductor-metal (MSM) structures , 1971 .

[15]  K Kyuma,et al.  Variable-sensitivity photodetector that uses a metal-semiconductor-metal structure for optical neural networks. , 1991, Optics letters.

[16]  Hirofumi Kan,et al.  Properties and design theory of ultrafast GaAs metal-semiconductor-metal photodetector with symmetrical Schottky contacts , 1990 .

[17]  S S Lee,et al.  Photodetector sensitivity control for weight setting in optoelectronic neural networks. , 1991, Applied optics.